Also, some news organizations continued to use an outdated estimate of 70,000 deaths, made once by the World Health Organization.

Accordingly, an estimate of about 1,000 deaths due to the removal is certainly not unreasonable.

A preliminary estimate of 20,500 similar deaths in 1985 (8.6 per 100,000) represents a drop of 14 percent.

The police doctor, who gave an estimate of 34 deaths, admitted that the exact figure would never be known.

The government earlier gave an estimate of 3,000 deaths.

In January, the agency's researchers corrected calculation errors and published a revised estimate of 365,000 deaths.

The estimate of 2.7 million annual deaths was the upper range provided by the genome researchers.

By 2006 upper estimates of 100,000 deaths were suggested.

The Government said its estimate of five million premature deaths was a conservative one based on current smoking rates.

It said survivors believe that 5,200 people have died in 18 villages alone, compared to an official estimate of 765 deaths in the entire region.